Traditional measures of active funds’ risk levels don’t capture crucial elements of skill or performance attribution, according to Harbour Asset Management portfolio manager, Shane Solly.
In a recently-penned article, Solly says the standard fund risk indicator – tracking error – is not up to the job of fully explaining how managers derive returns.
“Tracking error is a returns based measure, and is a proxy for systematic factor risks such as tilts to sectors, industries, macro trends and market timing,” he says in the article.
While tracking error provides useful information, Solly says investors must dig deeper in the data to understand the risk-taking and skill levels of fund managers.
He says comparing point-in-time fund holdings versus those of its benchmark – to create a measure called ‘active share’ – gives a better picture of “what exactly a manager is doing to drive performance, rather than drawing conclusions from observed returns”.
An ‘active share’ score of 0 per cent indicates a fund is matching its benchmark stock-for-stock in the same proportions. At the other end of the scale, a 100 per cent active share score shows a fund and its benchmark have no securities in common.
Solly told Investment News NZ that an active share approach would reveal ‘closet benchmark’ managers as well as those who use systemic market factors (for example, sector tilts) to generate returns.
“Today investors can use efficient, low-cost solutions to access some of those traditional market factors,” he said.
Solly said active share measures also flush out market timing managers.
“As an investor in an equity fund that’s holding 50 per cent cash, you have to be happy you’re leaving that market-timing decision to the manager,” he said.
Solly said while offshore researchers had used ‘active share’ metrics for a long time, over the last five years it had become much easier to “unbundle” the underlying fund strategies
He said some local research houses and asset consultants had also used ‘active share’ metrics for many years – albeit without making their findings explicit to investors.
The article cites US research showing managers with high active share scores tend to perform better over the long term. However, Solly also warns active share “is not a complete measure of risk”.
“Active Share does not account for portfolio construction, concentration of positions or diversification,” the article says.
But despite its limitations, Solly said a regularly-calculated and published active share measure would be a useful addition to the investor tool-kit.
The article says the ability to reveal fund risk changes over time would be “perhaps the key benefit” of an active share metric.
“Risk change over time, as measured by Active Share, could be part of a manager’s strategy,” the article says.
“But risk change could indicate style drift (for example becoming more index aware), behavioural biases (for example risk aversion) or excessive risk taking incentives. Any of these changes may deliver investment performance that is very different from what the fund has generated in the past.”